CN106815832A - A kind of steel mesh automatic image registration method and system of surface mounting technology - Google Patents
A kind of steel mesh automatic image registration method and system of surface mounting technology Download PDFInfo
- Publication number
- CN106815832A CN106815832A CN201611187643.7A CN201611187643A CN106815832A CN 106815832 A CN106815832 A CN 106815832A CN 201611187643 A CN201611187643 A CN 201611187643A CN 106815832 A CN106815832 A CN 106815832A
- Authority
- CN
- China
- Prior art keywords
- registration
- point
- steel mesh
- image
- rectangle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30141—Printed circuit board [PCB]
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of steel mesh automatic image registration method of surface mounting technology, belong to image procossing and field of optical measurements.The special type that the inventive method is not influenceed using the distance between image boundary rectangle and registration point by any putting position of steel mesh and angle, registration is carried out using registration point and design dot chart and steel mesh finished figure picture is designed, the calculating of transmission feature matrix is converted to the translation, rotation that by steel mesh and designed image calculated required for affine transformation is registering on time, the calculating for scaling 3 parameters, registering automatic test is carried out after registration, secondary registration is carried out after registration failure.Simultaneously The invention also achieves a kind of steel mesh automatic image registration system of surface mounting technology.Technical solution of the present invention is not influenceed by the putting position and angles of registering product, with registration accurately and fast and it is simple efficiently the characteristics of, while effectively raising registering success rate.
Description
Technical field
The invention belongs to image procossing and field of optical measurements, more particularly, to a kind of steel mesh of surface mounting technology
Automatic image registration method.
Background technology
As developing rapidly for electronic technology is also improving constantly with progress, the integrated level and complexity of PCB circuits, paster
Element has accounted for 50% to 90% in the overall component population of circuit board, and surface mounting technology (SMT) is current Electronic Assemblies
A kind of most popular technology and technique in industry.Steel mesh is a kind of SMT particular manufacturing crafts of laser cutting, and major function is side
Help the deposition of tin cream, it is therefore an objective to which the tin cream of accurate quantity is transferred to the accurate location on sky PCB.Now with processing technology with
And the development of electron trade is become better and better, the size of perforate is also less and less on SMT steel meshes, and quantity is more also into the increase of geometry
Up to thousands of or even up to ten thousand hole sizes from 0.1 micron to the distribution of several centimeters.It is typically all before this to be put by artificial use
Big mirror and other aids with the naked eye carry out the quality in observation detection steel mesh hole one by one, and artificial vision's detection is a large amount of by consumption
Time be difficult to ensure that the accuracy of detection simultaneously.Therefore the Automatic Measurement Technique based on machine vision will be following steel mesh detection
Inexorable trend.
Large format (more than 600*600mm) steel mesh is carried out by single area array CCD and telecentric lens cooperation motor machine platform
Snake scan be imaged, image by precision be spliced to form full width face image carried out by characteristic point with vector design document it is registering,
Call detection algorithm to carry out defects detection and obtain quality inspection result.It is automatic and accurate how steel mesh image is carried out with design document
Registration is the key before defects detection.Existing method for registering images usually needs the artificial characteristic point that clicks, and does not possess automatic
Property and high efficiency；Even if there is the method for autoregistration, existing autoregistration algorithm may be only available for steel mesh to be put and less inclines
Oblique and rotation.It is difficult to be applied to the stretcher strain in process of steel mesh random angles, steel mesh to registering success rate
Influence, and do not possess intelligent detection registration effect and Automatic Optimal adjustment is equipped with the functions such as raising success rate.
The content of the invention
In order to solve prior art above technological deficiency and technical problem, the invention provides a kind of surface mounting technology
Steel mesh automatic image registration method, matching characteristic point is chosen its object is to automatic；And the randomness with regard to steel mesh angles is carried
Rational usability methods are gone out；Add internal automatic test matching effect and secondary registration.Thus existing steel mesh is solved to match somebody with somebody
Accurate low, poor for applicability with the defect inspection method efficiency and inadequate technical problem such as intellectuality.
To achieve the above object, according to one aspect of the present invention, there is provided a kind of steel mesh image of surface mounting technology
Autoegistration method, the method is comprised the following steps：
(1) steel mesh design profile is obtained by steel mesh design document, asks for design profile minimum enclosed rectangle, obtain rectangle four
The coordinate on individual summit and the length of rectangle；
(2) area and barycentric coodinates of all intercommunicating pores in design profile are extracted, and design point is generated according to barycentric coodinates
System of battle formations picture；
(3) 4 design registration points are asked for by 4 summits and design dot matrix image of rectangle successively；
(4) finished product steel mesh high-contrast image is gathered, if finished product steel mesh is too big, collection steel mesh multiple topography is carried out
The complete steel mesh image of splicing composition, reusing iterative method carries out bilinear interpolation sub-pix Threshold segmentation treatment image；Ask for place
Image-region minimum enclosed rectangle and the coordinate on four summits of rectangle and the length of rectangle are calculated after reason；
(5) area and barycentric coodinates in the steel mesh image connectivity hole after extraction process, steel mesh is generated by all barycentric coodinates
Dot matrix image；
(6) 4 the first registration points are asked for by 4 summits of rectangle and steel mesh dot matrix image successively；
(7) found respectively in steel mesh dot matrix image registering as second away from 44 nearest focus points of the first registration point
Point；
(8) 3 the first registration points for choosing 3 design registration points and correspondence position carry out affine transformation, obtain feature and imitate
Matrix is penetrated, design profile is covered on steel mesh image after affine matrix is converted；
(9) autoregistration test is carried out to the design profile converted by affine matrix, judges whether registration success, if
Then registration terminates；The second registration point that failure registration point is replaced by correspondence position is otherwise continued into registration.
Further, registration point asks for process and is divided into following sub-step in the step (3) and step (6)：
(11) summit with rectangle is as the center of circle, with RdDrawn for radius and justified, wherein, Rd=kL, L are the length of rectangle, and k is
First radius multiple；The span of k is 5% ＜ k ＜ 20%, preferably k=10%；
(12) common factor of circle and dot matrix image is sought, judges whether there is focus point in common factor, having then obtain the center of gravity in occuring simultaneously
Coordinate, and perform step (34)；Otherwise perform step (33)；
(13) k=2k, updates radius RdAnd circle is repainted, perform step (32)；
(14) distance of all barycentric coodinates in asking rectangle summit and occuring simultaneously, takes the minimum barycentric coodinates of distance as registration
Point.
Further, in the step (4) iterative method carry out bilinear interpolation sub-pix Threshold segmentation treatment image it is specific
It is divided into following sub-step：
(41) each pixel of the transition portion between selection steel mesh display foreground and background carries out 10 × 10 interpolation；
(42) threshold value of steel mesh image is tried to achieve using iterative method, row threshold division is entered to the steel mesh image after interpolation.
Further, autoregistration test is divided into following sub-step in the step (9)：
(91) the connection hole area and barycentric coodinates of design profile after affine matrix is converted are extracted, is given birth to by all barycentric coodinates
Into registering dot matrix image；
(92) with 3 selected the first registration points one of them as the center of circle, with RcDrawn for radius and justified, wherein, Rc=jl, j are
Second radius multiple；The span of j is 1% ＜ j ＜ 10%, preferably j=3%；L is lie farthest away in 4 the first registration points
The distance between two registration points；The center of gravity point coordinates in circle and registering dot matrix image common factor is tried to achieve, the first common factor point set is designated as
Close；The center of gravity point coordinates in circle and steel mesh dot matrix image common factor is tried to achieve, the second common factor point set is designated as；
(93) contrast successively in the first common factor point set internal coordinate point and the second common factor point set correspondence position coordinate points it
Between distance, distance is designated as registering failed point more than the coordinate points of maximum allowable offset, if the number of registering failed point is more than matching somebody with somebody
Quasi- threshold value, then judge current the first registration point registration failure as the center of circle, and be designated as unsuccessfully registration point；Otherwise judge current work
It is the first registration point registration success in the center of circle；Wherein, registering threshold value span is common factor point set internal coordinate point sum
30% to 50%, preferably the 40% of common factor point set internal coordinate point sum；The span of maximum allowable offset is arrived for 50um
150um, preferably 100um；
(94) repeat step (92) and (93) are tested until the first registration point selected by 3 all completes autoregistration, if 3 institutes
Select all registration successes of the first registration point, then registration success；Otherwise registration failure.
It is another aspect of this invention to provide that there is provided a kind of steel mesh automatic image registration system of surface mounting technology, should
System includes following part：
Design profile boundary rectangle extraction module, for obtaining steel mesh design profile by steel mesh design document, asks for design
Profile minimum enclosed rectangle, obtains the coordinate on four summits of rectangle and the length of rectangle；
Design dot matrix image extraction module, area and barycentric coodinates for extracting all intercommunicating pores in design profile, and
According to barycentric coodinates generation design dot matrix image；
Design registration point extraction module, for being asked for 4 designs and being matched somebody with somebody by 4 summits and design dot matrix image of rectangle successively
On schedule；
Steel mesh image boundary rectangle extraction module, for gathering finished product steel mesh high-contrast image, if finished product steel mesh is too big,
Then collection steel mesh multiple topography carries out the splicing complete steel mesh image of composition, and reusing iterative method carries out bilinear interpolation Asia picture
Plain Threshold segmentation processes image；Ask for after treatment image-region minimum enclosed rectangle and calculate the coordinate and square on four summits of rectangle
The length of shape；
Steel mesh dot matrix image extraction module, for the area and barycentric coodinates in the steel mesh image connectivity hole after extraction process,
Steel mesh dot matrix image is generated by all barycentric coodinates；
First registration point extraction module, matches somebody with somebody for asking for 4 first by 4 summits of rectangle and steel mesh dot matrix image successively
On schedule；
Second registration point extraction module, for found respectively in steel mesh dot matrix image away from 4 the first registration points it is nearest 4
Individual focus point is used as the second registration point；
Registration module, 3 the first registration points for choosing 3 design registration points and correspondence position carry out affine transformation,
Feature affine matrix is obtained, design profile is covered on steel mesh image after affine matrix is converted；
Registering test module, for carrying out autoregistration test to the design profile converted by affine matrix, judgement is
No registration success, if then registration terminates；The second registration point that failure registration point is replaced by correspondence position is otherwise continued into registration.
Further, registration point asks for process point in the design registration point extraction module and the first registration point extraction module
It is following submodule：
Dumpling made of glutinous rice flour module is drawn, for a summit of rectangle as the center of circle, with RdDrawn for radius and justified, wherein, Rd=kL, L are square
The length of shape, k is the first radius multiple；The span of k is 5% ＜ k ＜ 20%, preferably k=10%；
Whether seek common ground submodule, the common factor for asking circle and dot matrix image, judges there is focus point in common factor, has, and obtains
Barycentric coodinates in common factor, and perform step (34)；Otherwise perform step (33)；
Dumpling made of glutinous rice flour module is updated, for updating k=2k, radius R is updateddAnd circle is repainted, perform step (32)；
Registration point submodule is sought, the distance for seeking all barycentric coodinates in rectangle summit and common factor takes distance minimum
Barycentric coodinates are used as registration point.
Further, iterative method carries out bilinear interpolation sub-pix threshold value in the steel mesh image boundary rectangle extraction module
Dividing processing image is specifically divided into following submodule：
Interpolation submodule, each pixel for selecting the transition portion between steel mesh display foreground and background carries out 10 ×
10 interpolation；
Threshold segmentation submodule, the threshold value for being tried to achieve steel mesh image using iterative method is entered to the steel mesh image after interpolation
Row threshold division.
Further, autoregistration test is divided into following submodule in the registering test module：
Registering dot matrix image generates submodule, connection hole area and weight for extracting design profile after affine matrix is converted
Heart coordinate, registering dot matrix image is generated by all barycentric coodinates；
Registering common factor submodule, for 3 selected the first registration points one of them as the center of circle, with RcFor radius is drawn
Circle, wherein, Rc=jl, j are the second radius multiple；The span of j is 1% ＜ j ＜ 10%, preferably j=3%；L is 4 the
The distance between two registration points of lie farthest away in one registration point；The focus point tried to achieve in circle and registering dot matrix image common factor is sat
Mark, is designated as the first common factor point set；The center of gravity point coordinates in circle and steel mesh dot matrix image common factor is tried to achieve, the second common factor point set is designated as
Close；
Registration point judging submodule, for contrasting successively in the first common factor point set internal coordinate point and the second common factor point set
The distance between correspondence position coordinate points, distance is designated as registering failed point more than the coordinate points of maximum allowable offset, if registration loses
The number for losing a little is more than registering threshold value, then judge current the first registration point registration failure as the center of circle, and be designated as unsuccessfully registration
Point；Otherwise judge current the first registration point registration success as the center of circle；Wherein, registering threshold value span is all intersection points
30% to the 50% of sum, the 40% of preferably all intersection point sums；The span of maximum allowable offset is arrived for 50um
150um, preferably 100um；
Registering judging submodule, for repeat step registration common factor submodule and registration point judging submodule until 3 institutes
Select the first registration point all to complete autoregistration to test, if all registration successes of the first registration point selected by 3, registration success；Otherwise
Registration failure.
In general, by the contemplated above technical scheme of the present invention compared with prior art, it is special with following technology
Levy and beneficial effect：
(1) minimum enclosed rectangle of the utilization figure of the invention can with arbitrary putting position and angle phase
The change answered, but registration features point and the constant property feature of the absolute minimum range in rectangle summit, can effectively solve arbitrarily rotation
Turn the influence to registration effect.
(2) present invention cleverly make use of splitting with computing and the automatic hunting zone that increases between image and image
Characteristic area greatly reduces the amount of calculation and complexity for finding characteristic point.By using iterative method selected threshold and bilinearity
The segmentation of interpolation sub-pix ensure that the precision and accuracy of detection afterwards of registration.Can be than classics on the premise of the degree of accuracy is ensured
Sub-pixel Edge Detection faster completes contours extract.
(3) technical solution of the present invention in view of actual steel mesh there may be in process stretching and heating power deformation and
Other are difficult to the influence expected, by automatic test and the mechanism of adjust automatically registration point and secondary registration, compared to existing
Technology is registering to be had higher success rate.
Brief description of the drawings
Fig. 1 is the inventive method flow chart；
Fig. 2 design profiles file and its minimum enclosed rectangle and 4 summit schematic diagrames；
Fig. 3 is the minimum range registration point schematic diagram for searching that occured simultaneously as the center of circle and dot chart with the 1st summit；
Fig. 4 is steel mesh image sub-pix Threshold segmentation schematic diagram；
Fig. 5 is the steel mesh image and its minimum enclosed rectangle and 4 summit schematic diagrames arbitrarily put；
Fig. 6 is the first registration point and potential interference the second registration point schematic diagram in image；
Fig. 7 be steel mesh image it is registering with file after partial schematic diagram；
Fig. 8 is steel mesh image failure schematic diagram registering with file；
Fig. 9 is that registration is unsuccessfully tested success schematic diagram after secondary registration automatically；
Figure 10 be steel mesh image it is registering with design document after exist hole position skew schematic diagram.
Specific embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.As long as additionally, technical characteristic involved in invention described below each implementation method
Not constituting conflict each other can just be mutually combined.
As shown in figure 1, the inventive method flow is as follows：
(1) information such as steel mesh processing Gerber files, parsing hole site, shape size are read and obtains steel mesh contour vector text
Part simultaneously shows as shown in Figure 2；
Ask for the minimum enclosed rectangle of whole file area and be calculated rectangle upper left, upper right, lower-left, the top of bottom right four
The coordinate of point and the length of rectangle, as shown in rectangle frame in Fig. 2 and 4 round dots.The step of minimum enclosed rectangle for asking for image is：
The starting point coordinate (x1, y1) and terminal point coordinate (x2, y2) of each row searching each column target of scan image are obtained in this row successively
Barycentric coodinates be (x, y), calculate each row barycentric coodinates (xi, yi) and carry out fitting a straight line using least square method and obtain level master
Axle linear equation y=k1x+b1.Row center of gravity and vertical major y=k2x+b2 are similarly asked each row.Real center of gravity is level master
The intersection point of axle and vertical major.The constant downward translation in horizontal spindle direction is kept to find whether straight line intersects with target, last
Secondary intersecting linear position as boundary rectangle lower boundary.Similarly can be in the hope of the upper and right boundary of boundary rectangle as first
Beginning boundary rectangle.Then by certain direction of rotation and anglec of rotation interval rotation initial rectangular, the minimum external square of area is found
Shape is used as preferred target.The mutually perpendicular four edges of rectangle are represented with linear equation, and the intersection point of straight line is then external square
4 summits of shape.
(2) to all connection bore regions of design profile file according to asking the method for center of gravity to extract intercommunicating pore in step (1)
Area and center of gravity coordinate, and barycentric coodinates point is generated width design dot chart, the dot in such as Fig. 3 show design
Dot chart.
(3) 4 summits with rectangle are as the center of circle, and radius initial value is rectangle 10% picture circle long, and friendship is asked with design dot chart
Collection, and the barycentric coodinates point in being occured simultaneously.If do not occured simultaneously, one times of radius of circle automatic increase, the picture circle for repeating asks friendship
Collection in common factor untill having focus point.Ask for occuring simultaneously the distance of interior barycentric coodinates and current rectangle summit, and take minimum range
Corresponding coordinate points are a registration features point.4 registration points nearest with 4 vertex distances of rectangle are obtained, design is designated as
Registration point.As shown in figure 3, quadrant is the circle drawn as the center of circle with top left corner apex, the point corresponding to fork is beaten in figure
It is the registration point closest with top left corner apex.
(4) bilinear interpolation sub-pix threshold value point is carried out using iterative method to the high-contrast steel mesh image that collection has been spliced
Cut as shown in Figure 4.Transition portion between selection steel mesh display foreground and background enters row interpolation.Thus each pixel is carried out
The interpolation of 10*10, the threshold value calculated using iterative method enters row threshold division to the image after interpolation.Ask for whole image region
Interior minimum enclosed rectangle and the length of the coordinate and rectangle that are calculated four summits of rectangle.Such as rectangle frame and 4 round dots in Fig. 5
It is shown,
(5) area and the coordinate of center of gravity in the image zooming-out hole after splitting to steel mesh sub-pix, all barycentric coodinates points
The image of a width dot matrix is generated, steel mesh dot chart is designated as, as shown in Fig. 6 dots.
(6) 4 summits of image boundary rectangle are the center of circle, and radius initial value is rectangle 10% picture circle long, with steel mesh dot matrix
Figure seeks common ground, and the barycentric coodinates point in being occured simultaneously.If do not occured simultaneously, one times of radius of circle automatic increase, the picture for repeating
Circle seeks common ground untill having common factor.Ask for occuring simultaneously the distance of interior barycentric coodinates and current rectangle summit, and take minimum range institute
Corresponding coordinate points are a registration features point.Obtain 4 first registration points nearest with 4 vertex distances of rectangle.As in Fig. 6
The point beaten corresponding to fork is the registration point closest with top left corner apex.Steel mesh is put in the presence of randomness, and angle is inclined and is difficult to
Avoid, the minimum enclosed rectangle of image also can with the inclined degree put respective change, but with rectangle vertex distance most
What the absolute property of small feature focus point was no variation in, thus the method goes for the random inclination of angles.
(7) understand because heat expansion is cold as the number and closeness of open cell content increase during cutting processing in view of steel mesh
Contracting and stretching cause a certain degree of deformation of global shape.If having neighbor point and rectangle vertex distance near the first registration point
With the first registration point minimum range closely, then backed up using neighbor point as potential interference point, as possible second
Registration point.As what circle in Fig. 6 was irised out is backed up as potential interference point.
(8) find out 3 design registration points carries out affine transformation with 3 coordinates of the first registration point of correspondence position, draws
Feature affine matrix.Design profile file by affine matrix convert after displacement, the anglec of rotation and scaling all with steel mesh figure
As consistent and coordinate unification.
(9) registration result is tested automatically.To the face of the design profile file drawing holes after step (8) affine transformation
The coordinate of product and center of gravity, generates all barycentric coodinates points the image of the new dot matrix of one width, is designated as registering dot chart.With step
(8) 3 the first registration points selected in are the center of circle, and the 3% a length of radius of rectangle hypotenuse draws circle, tries to achieve circle and registering dot matrix image
Center of gravity point coordinates in common factor, is designated as the first common factor point set；The center of gravity point coordinates in circle and steel mesh dot matrix image common factor is tried to achieve,
It is designated as the second common factor point set；Correspondence position in the first common factor point set internal coordinate point and the second common factor point set is contrasted successively to sit
The distance between punctuate, coordinate points of the distance more than 100um are designated as registering failed point, if the number of registering failed point is more than common factor
The 40% of point set internal coordinate point sum, then judge currently to fail as the first registration point registration in the center of circle, and be designated as failure and match somebody with somebody
On schedule；Otherwise judge current the first registration point registration success as the center of circle；The first registration point all completes autoregistration selected by 3
After test, if all registration successes of the first registration point selected by 3, registration success, as shown in Figure 8；If failure, as shown in fig. 7, then
First registration point of registration failure is replaced by after the second registration point and repeats autoregistration test continuation registration.Images after registration office
Portion by registration result as shown in figure 9, substantially can see that steel mesh makes defect.
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention, it is all in essence of the invention
Any modification, equivalent and improvement made within god and principle etc., should be included within the scope of the present invention.
Claims (8)
1. a kind of steel mesh automatic image registration method of surface mounting technology, it is characterised in that the method is comprised the following steps：
(1) steel mesh design profile is obtained by steel mesh design document, asks for design profile minimum enclosed rectangle, obtain four tops of rectangle
The coordinate of point and the length of rectangle；
(2) area and barycentric coodinates of all intercommunicating pores in design profile are extracted, and according to barycentric coodinates generation design dot chart
Picture；
(3) 4 design registration points are asked for by 4 summits and design dot matrix image of rectangle successively；
(4) finished product steel mesh high-contrast image is gathered, reusing iterative method carries out the treatment of bilinear interpolation sub-pix Threshold segmentation
Image；Image-region minimum enclosed rectangle and the coordinate on four summits of rectangle and the length of rectangle are calculated after asking for treatment；
(5) area and barycentric coodinates in the steel mesh image connectivity hole after extraction process, steel mesh dot matrix is generated by all barycentric coodinates
Image；
(6) 4 the first registration points are asked for by 4 summits of rectangle and steel mesh dot matrix image successively；
(7) found respectively in steel mesh dot matrix image away from 4 nearest focus points of 4 the first registration points as the second registration point；
(8) 3 the first registration points for choosing 3 design registration points and correspondence position carry out affine transformation, obtain the affine square of feature
Battle array, design profile is covered on steel mesh image after affine matrix is converted；
(9) autoregistration test is carried out to the design profile converted by affine matrix, judges whether registration success, if then matching somebody with somebody
Standard terminates；The second registration point that failure registration point is replaced by correspondence position is otherwise continued into registration.
2. a kind of steel mesh automatic image registration method of surface mounting technology according to claim 1, it is characterised in that institute
State registration point in step (3) and step (6) and ask for process and be divided into following sub-step：
(11) summit with rectangle is as the center of circle, with RdDrawn for radius and justified, wherein, Rd=kL, L are the length of rectangle, and k is first
Radius multiple；
(12) common factor of circle and dot matrix image is sought, judges whether there is focus point in common factor, having then obtain the barycentric coodinates in occuring simultaneously,
And perform step (34)；Otherwise perform step (33)；
(13) k=2k, updates radius RdAnd circle is repainted, perform step (32)；
(14) distance of all barycentric coodinates in asking rectangle summit and occuring simultaneously, takes the minimum barycentric coodinates of distance as registration point.
3. a kind of steel mesh automatic image registration method of surface mounting technology according to claim 1, it is characterised in that institute
State iterative method in step (4) and carry out bilinear interpolation sub-pix Threshold segmentation treatment image and be specifically divided into following sub-step：
(41) each pixel of the transition portion between selection steel mesh display foreground and background carries out 10 × 10 interpolation；
(42) threshold value of steel mesh image is tried to achieve using iterative method, row threshold division is entered to the steel mesh image after interpolation.
4. a kind of steel mesh automatic image registration method of surface mounting technology according to claim 1, it is characterised in that institute
Autoregistration test is divided into following sub-step in stating step (9)：
(91) the connection hole area and barycentric coodinates of design profile after affine matrix is converted are extracted, is matched somebody with somebody by the generation of all barycentric coodinates
Quasi- dot matrix image；
(92) with 3 selected the first registration points one of them as the center of circle, with RcDrawn for radius and justified, wherein, Rc=jl, j are second
Radius multiple；L is the distance between two registration points of lie farthest away in 4 the first registration points；Try to achieve circle and registering dot matrix image
Center of gravity point coordinates in common factor, is designated as the first common factor point set；The center of gravity point coordinates in circle and steel mesh dot matrix image common factor is tried to achieve,
It is designated as the second common factor point set；
(93) contrast successively in the first common factor point set internal coordinate point and the second common factor point set between correspondence position coordinate points
Distance, distance is designated as registering failed point more than the coordinate points of maximum allowable offset, if the number of registering failed point is more than registering threshold
Value, then judge current the first registration point registration failure as the center of circle, and be designated as unsuccessfully registration point；Otherwise judge current as circle
The first registration point registration success of the heart；
(94) repeat step (92) and (93) all complete autoregistration test until the first registration point selected by 3, if the selected by 3
The all registration successes of one registration point, then registration success；Otherwise registration failure.
5. the steel mesh automatic image registration system of a kind of surface mounting technology, it is characterised in that the system includes following part：
Design profile boundary rectangle extraction module, for obtaining steel mesh design profile by steel mesh design document, asks for design profile
Minimum enclosed rectangle, obtains the coordinate on four summits of rectangle and the length of rectangle；
Design dot matrix image extraction module, area and barycentric coodinates for extracting all intercommunicating pores in design profile, and according to
Barycentric coodinates generation design dot matrix image；
Design registration point extraction module, for asking for 4 design registrations by 4 summits and design dot matrix image of rectangle successively
Point；
Steel mesh image boundary rectangle extraction module, for gathering finished product steel mesh high-contrast image, reuse iterative method carry out it is double
Linear interpolation sub-pix Threshold segmentation processes image；Ask for after treatment image-region minimum enclosed rectangle and calculate four, rectangle top
The coordinate of point and the length of rectangle；
Steel mesh dot matrix image extraction module, for the area and barycentric coodinates in the steel mesh image connectivity hole after extraction process, by institute
There are barycentric coodinates to generate steel mesh dot matrix image；
First registration point extraction module, for asking for 4 first registrations by 4 summits of rectangle and steel mesh dot matrix image successively
Point；
Second registration point extraction module, for finding 4 weights nearest away from 4 the first registration points respectively in steel mesh dot matrix image
The heart o'clock is used as the second registration point；
Registration module, 3 the first registration points for choosing 3 design registration points and correspondence position carry out affine transformation, obtain
Feature affine matrix, design profile is covered on steel mesh image after affine matrix is converted；
Registering test module, for carrying out autoregistration test to the design profile converted by affine matrix, judges whether to match somebody with somebody
Quasi- success, if then registration terminates；The second registration point that failure registration point is replaced by correspondence position is otherwise continued into registration.
6. a kind of steel mesh automatic image registration system of surface mounting technology according to claim 5, it is characterised in that institute
Registration point asks for process and is divided into following submodule in stating design registration point extraction module and the first registration point extraction module：
Dumpling made of glutinous rice flour module is drawn, for a summit of rectangle as the center of circle, with RdDrawn for radius and justified, wherein, Rd=kL, L are rectangle
Long, k is the first radius multiple；
Whether seek common ground submodule, the common factor for asking circle and dot matrix image, judges there is focus point in common factor, has then to obtain and occurs simultaneously
Interior barycentric coodinates, and perform step (34)；Otherwise perform step (33)；
Dumpling made of glutinous rice flour module is updated, for updating k=2k, radius R is updateddAnd circle is repainted, perform step (32)；
Registration point submodule is sought, the distance for seeking all barycentric coodinates in rectangle summit and common factor takes the minimum center of gravity of distance
Coordinate is used as registration point.
7. a kind of steel mesh automatic image registration system of surface mounting technology according to claim 5, it is characterised in that institute
State iterative method in steel mesh image boundary rectangle extraction module and carry out bilinear interpolation sub-pix Threshold segmentation treatment image and specifically divide
It is following submodule：
Interpolation submodule, each pixel for selecting the transition portion between steel mesh display foreground and background carries out 10 × 10
Interpolation；
Threshold segmentation submodule, the threshold value for trying to achieve steel mesh image using iterative method carries out threshold to the steel mesh image after interpolation
Value segmentation.
8. a kind of steel mesh automatic image registration system of surface mounting technology according to claim 5, it is characterised in that institute
Autoregistration test is divided into following submodule in stating registering test module：
Registering dot matrix image generates submodule, and the connection hole area and center of gravity for extracting design profile after affine matrix is converted are sat
Mark, registering dot matrix image is generated by all barycentric coodinates；
Registering common factor submodule, for 3 selected the first registration points one of them as the center of circle, with RcDrawn for radius and justified, its
In, Rc=jl, j are the second radius multiple；L is the distance between two registration points of lie farthest away in 4 the first registration points；Try to achieve
Center of gravity point coordinates in circle and registering dot matrix image common factor, is designated as the first common factor point set；Try to achieve circle and steel mesh dot matrix image is handed over
Center of gravity point coordinates in collection, is designated as the second common factor point set；
Registration point judging submodule, for contrasting correspondence in the first common factor point set internal coordinate point and the second common factor point set successively
The distance between position coordinates point, distance is designated as registering failed point more than the coordinate points of maximum allowable offset, if registering failed point
Number be more than registering threshold value, then judge current the first registration point registration failure as the center of circle, and be designated as unsuccessfully registration point；It is no
Then judge current the first registration point registration success as the center of circle；
Registering judging submodule, for repeating registering common factor submodule and registration point judging submodule until first matching somebody with somebody selected by 3
Autoregistration test is all completed on schedule, if all registration successes of the first registration point selected by 3, registration success；Otherwise registration failure.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611187643.7A CN106815832B (en) | 2016-12-20 | 2016-12-20 | A kind of steel mesh automatic image registration method and system of surface mounting technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611187643.7A CN106815832B (en) | 2016-12-20 | 2016-12-20 | A kind of steel mesh automatic image registration method and system of surface mounting technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106815832A true CN106815832A (en) | 2017-06-09 |
| CN106815832B CN106815832B (en) | 2019-05-21 |
Family
ID=59109099
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611187643.7A Active CN106815832B (en) | 2016-12-20 | 2016-12-20 | A kind of steel mesh automatic image registration method and system of surface mounting technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106815832B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416773A (en) * | 2018-03-08 | 2018-08-17 | 凌云光技术集团有限责任公司 | A kind of the PCB aperture method of inspection and system of the over-determined systems based on registration point |
| CN108986181A (en) * | 2018-06-15 | 2018-12-11 | 广东数相智能科技有限公司 | Image processing method, device and computer readable storage medium based on dot |
| WO2021077759A1 (en) * | 2019-10-21 | 2021-04-29 | 推想医疗科技股份有限公司 | Image matching method, apparatus and device, and storage medium |
| CN113242652A (en) * | 2021-05-20 | 2021-08-10 | 上海望友信息科技有限公司 | Spraying graph generation method and system, electronic equipment and storage medium |
| CN113888558A (en) * | 2021-09-26 | 2022-01-04 | 武汉钢铁有限公司 | Ladle slag surface image correction method |
| CN114429416A (en) * | 2022-02-21 | 2022-05-03 | 合肥龙旗智能科技有限公司 | Processing method, device, system and equipment for attaching dial and bezel |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177444A (en) * | 2013-03-08 | 2013-06-26 | 中国电子科技集团公司第十四研究所 | Automatic SAR (synthetic-aperture radar) image rectification method |
| CN103778626A (en) * | 2013-12-31 | 2014-05-07 | 北京理工大学 | Quick image registration method based on visual remarkable area |
| CN103854276A (en) * | 2012-12-04 | 2014-06-11 | 株式会社东芝 | Image registration device and method, image segmentation device and method and medical image device |
| CN104318568A (en) * | 2014-10-24 | 2015-01-28 | 武汉华目信息技术有限责任公司 | Method and system for image registration |
| CN104463866A (en) * | 2014-12-04 | 2015-03-25 | 无锡日联科技有限公司 | Local shape matching method based on outline random sampling |
| CN105389815A (en) * | 2015-10-29 | 2016-03-09 | 武汉联影医疗科技有限公司 | Breast image registration method and device |
-
2016
- 2016-12-20 CN CN201611187643.7A patent/CN106815832B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103854276A (en) * | 2012-12-04 | 2014-06-11 | 株式会社东芝 | Image registration device and method, image segmentation device and method and medical image device |
| CN103177444A (en) * | 2013-03-08 | 2013-06-26 | 中国电子科技集团公司第十四研究所 | Automatic SAR (synthetic-aperture radar) image rectification method |
| CN103778626A (en) * | 2013-12-31 | 2014-05-07 | 北京理工大学 | Quick image registration method based on visual remarkable area |
| CN104318568A (en) * | 2014-10-24 | 2015-01-28 | 武汉华目信息技术有限责任公司 | Method and system for image registration |
| CN104463866A (en) * | 2014-12-04 | 2015-03-25 | 无锡日联科技有限公司 | Local shape matching method based on outline random sampling |
| CN105389815A (en) * | 2015-10-29 | 2016-03-09 | 武汉联影医疗科技有限公司 | Breast image registration method and device |
Non-Patent Citations (2)
| Title |
|---|
| 刘今越 等: "基于机器视觉的SMT钢网质量检测技术研究", 《仪器仪表学报》 * |
| 杨林杰: "基于轮廓特征的目标匹配算法研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108416773A (en) * | 2018-03-08 | 2018-08-17 | 凌云光技术集团有限责任公司 | A kind of the PCB aperture method of inspection and system of the over-determined systems based on registration point |
| CN108986181A (en) * | 2018-06-15 | 2018-12-11 | 广东数相智能科技有限公司 | Image processing method, device and computer readable storage medium based on dot |
| WO2021077759A1 (en) * | 2019-10-21 | 2021-04-29 | 推想医疗科技股份有限公司 | Image matching method, apparatus and device, and storage medium |
| US11954860B2 (en) | 2019-10-21 | 2024-04-09 | Infervision Medical Technology Co., Ltd. | Image matching method and device, and storage medium |
| CN113242652A (en) * | 2021-05-20 | 2021-08-10 | 上海望友信息科技有限公司 | Spraying graph generation method and system, electronic equipment and storage medium |
| CN113888558A (en) * | 2021-09-26 | 2022-01-04 | 武汉钢铁有限公司 | Ladle slag surface image correction method |
| CN113888558B (en) * | 2021-09-26 | 2024-05-28 | 武汉钢铁有限公司 | Ladle slag surface image correction method |
| CN114429416A (en) * | 2022-02-21 | 2022-05-03 | 合肥龙旗智能科技有限公司 | Processing method, device, system and equipment for attaching dial and bezel |
| CN114429416B (en) * | 2022-02-21 | 2024-06-21 | 合肥龙旗智能科技有限公司 | Method, device, system and equipment for processing attachment of dial plate and bezel |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106815832B (en) | 2019-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106815832A (en) | A kind of steel mesh automatic image registration method and system of surface mounting technology | |
| CN113688807B (en) | Self-adaptive defect detection method, device, recognition system and storage medium | |
| CN106340044B (en) | Join automatic calibration method and caliberating device outside video camera | |
| CN106709947A (en) | RGBD camera-based three-dimensional human body rapid modeling system | |
| CN107170008B (en) | Depth map creating method and system and image blurring method and system | |
| CN114897864B (en) | Workpiece detection and defect judgment method based on digital-analog information | |
| CN107967713A (en) | Construction three-dimensional model building method and system based on spatial point cloud data | |
| CN104751458B (en) | A kind of demarcation angular-point detection method based on 180 ° of rotation operators | |
| CN101493942B (en) | Level set image segmentation method based on spacing moment | |
| CN107369204B (en) | Method for recovering basic three-dimensional structure of scene from single photo | |
| CN111354047B (en) | Computer vision-based camera module positioning method and system | |
| CN115170669A (en) | Identification and positioning method and system based on edge feature point set registration and storage medium | |
| CN105354816B (en) | Electronic component positioning method and device | |
| CN114419045A (en) | Method, device and equipment for detecting defects of photoetching mask plate and readable storage medium | |
| CN109365877A (en) | A kind of NC Machines processing automatic positioning method | |
| CN109255792A (en) | A kind of dividing method of video image, device, terminal device and storage medium | |
| CN103236056B (en) | Based on the image partition method of template matches | |
| CN106204564A (en) | A kind of laser photocentre extracting method | |
| CN116740099B (en) | OCT image segmentation method and device | |
| CN114359276B (en) | Steel die blanking optimization scheme obtaining method based on pockmark defects | |
| CN116542926A (en) | Method, device, equipment and storage medium for identifying defects of two-dimension codes of battery | |
| CN115239801A (en) | Object positioning method and device | |
| CN112819955B (en) | Improved three-dimensional model reconstruction method based on digital image | |
| CN115272180A (en) | Method for acquiring production parameters of printed circuit board, optical detection equipment, electronic equipment and storage medium | |
| CN109215049B (en) | Roof segmentation method, system and equipment based on multi-scale three-dimensional prior information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |